Prospective validation of the Good Outcome Following Attempted Resuscitation (GO-FAR) score for in-hospital cardiac arrest prognosis

Abstract

AimWe aimed to prospectively validate the Good Outcome Following Attempted Resuscitation (GO-FAR) score, which predicts the likelihood of survival to discharge neurologically intact or with minimal deficits (conscious, alert, and able to work) after in-hospital cardiac arrest (IHCA). MethodsInpatients experiencing an index episode of IHCA between 2010 and 2016 in hospitals participating in the Get With the Guidelines® — Resuscitation (GWTG-R) Registry were included. The score’s performance was prospectively validated in both all GWTG-R hospitals and in a subset of hospitals not part of the GWTG-R registry when the score was originally developed using prospective data. Score performance was stratified by hospital size, presence of residency training programs, and type of hospital ownership. Discrimination was measured by the c-statistic, calibration using a Hosmer-Lemeshow plot, and classification accuracy by the survival rates in each risk group. ResultsA total of 62,131 inpatients in 386 hospital were included. The GO-FAR score had similar discrimination (c-statistic 0.75, 95% CI 0.748−0.758), calibration, and classification accuracy as in the original study. Survival rates were somewhat higher due to a secular increase in survival of IHCA. In hospitals that were not part of the derivation population, the score performed as well as in the hospitals used to derive the score (c-statistic 0.75). The score performed similarly in hospitals of different sizes (c-statistic of 0.80 and 0.75 for hospital with ≤100 and >100 beds, respectively), with and without residency training programs (c-statistics of 0.76 and 0.75, respectively), and with different ownership structures (c-statistic of 0.79 for private, 0.74 for military government, and 0.76 for nonprofit hospital). ConclusionsThe GO-FAR score accurately classifies patients into risk groups based on their likelihood of survival to discharge with a good neurologic outcome following an episode of IHCA. Recalibration may be necessary using different point score cutoffs as IHCA survival increases.